Backtesting
A portfolio backtesting tool built on real crises.
Anyone can show you a backtest of the last bull market. MEDGE replays your exact allocation through COVID, Lehman, the taper tantrum and the 2022 rate shock — the windows where portfolios actually fail — and returns the full risk picture: CAGR, Sharpe, Sortino, max drawdown, CVaR and a base-100 equity curve.
How it works
1. Build
Paste tickers and weights, or start from 24 pre-built strategies. Long, short, multi-asset — no brokerage connection required.
2. Pick a window
A crisis preset (COVID, Lehman, taper tantrum, 2022) or any custom date range. The engine aligns every holding on the same trading days.
3. Read the risk
Equity curve, drawdown profile and the full metric table — plus a side-by-side against 24 preset portfolios on the identical window.
Try the crisis replay
Toggle period + strategy, see the curve redraw
The crisis-preset library
COVID Crash
Feb – Apr 2020The fastest 30% drawdown in history and the V-shaped rebound that punished panic sellers.
Lehman / GFC
2008 – 2009The systemic credit crisis. The stress test every 60/40 claim should survive.
Taper Tantrum
May – Sep 2013Rates repricing that hit bonds and EM at once — a correlation-breakdown case study.
2022 Rate Shock
Jan – Oct 2022Stocks AND bonds down double digits together. The year that broke naive diversification.
These are four of the 15 dated crisis and rally windows in the library. Custom windows work too — any start and end date — and the Stress Lab tests scenarios that never happened.
Beyond the equity curve
Real rebalancing, real costs
Five policies — none, monthly, quarterly, yearly, threshold — with transaction costs, weights that drift between rebalances, and all policies compared on the same holdings.
Why it happened
Exact per-holding attribution for any window, the benchmark-relative picture (beta, alpha, capture ratios) and how each holding behaved in bear, neutral and bull markets.
What if
Remove, replace or re-weight a holding, or tilt its return, and replay the whole backtest with the same rules — plus the effect of dropping each holding or adding five points to it.
What next
The Stress Lab: set shocks on indices, sectors, countries, rates, commodities or currencies and see every holding respond, with the historical episodes that looked most alike.
Beyond history: Monte Carlo
History only ran once. MEDGE’s Monte Carlo engine simulates thousands of alternative paths with three methods — Bootstrap, Geometric Brownian Motion and t-copula — so you can see the probability fan around your plan, not just the single path that happened. Metrics like max drawdown and CVaR are computed across the whole distribution. The methodology is documented openly on the methodology page.
Frequently asked questions
- Is the backtesting tool free?
- Yes. The Free tier, with no time limit, includes the whole backtesting lab — rebalancing policies with costs, Period Analysis with per-holding attribution, Regime Summary, What-If, External Ticker Shock, Weak & Strong Spots — plus Monte Carlo. Essential (€14.99/month) adds the Stress Lab, full Analysis, optimization on 12 objectives, Compare, the report and up to 10 saved portfolios (Free saves 1); Plus (€39.99/month, 7-day trial) adds measured exposure and macro risks.
- Do I need to connect a brokerage account?
- No. You paste tickers and weights (or start from one of 24 pre-built strategy templates). MEDGE never asks for brokerage credentials — portfolio analysis should not require sharing account access with a third party.
- How far back does the price history go?
- Coverage varies by instrument; most large US and European ETFs and stocks have 10+ years of daily closes, enough to replay 2008. Each backtest reports exactly which tickers had data for the window you chose, so silently truncated histories can’t skew results unnoticed.
- What metrics does a backtest return?
- Period return, CAGR, volatility, Sharpe, Sortino, max drawdown, Calmar and CVaR, a base-100 equity curve and an underwater drawdown chart — plus beta, alpha, tracking error, information ratio and up/down capture against the benchmark you choose, a monthly heatmap, the return distribution and the best and worst stretches.
- Does the backtest rebalance, and does it charge costs?
- Yes. Weights drift with prices between rebalances, and you choose the policy — none, monthly, quarterly, yearly or a threshold band — and a transaction cost from 0 to 500 basis points (0 by default). The same holdings are replayed under every policy side by side, so you can see what rebalancing actually cost or earned.
- Can I see why the portfolio made or lost money?
- Yes. For any window, the gain or loss is attributed exactly to each holding, with weights drifting and rebalancing as in the backtest — contributions plus costs add up to the portfolio return. Regime Summary shows how each holding behaved in bear, neutral and bull markets and in calm or turbulent ones.
- Can I compare my portfolio against standard strategies?
- Yes — one click benchmarks your portfolio against 24 preset allocations (60/40, All Weather, Permanent Portfolio, Golden Butterfly, Three-Fund, Yale endowment and more) over the same window, with a metric-by-metric strengths and weaknesses table.
Related reading
Find out how your portfolio handles 2008.
Free tier, no brokerage connection, results in under a minute.
Start backtesting